Some experienced developers of hard real-time embedded defense systems have complained about problems with the Ada programming language. These problems are real, but resourceful application builders and Ada language implementors are finding effective ways to work around them. Unfortunately, these solutions are based on special features of particular Ada compilation systems. They may involve subtle compiler-dependent application code, and expensive modifications to the compiler or Ada runtime system to fit a particular application. Whether we can solve the problems of programming real- time systems in Ada is therefore not so much an issues as whether we can provide economical, and preferably standard, Ada solutions to these problems. Modification to the Ada standard is not a sufficient solution, especially in the short term. Despite the benefits of commonality, it is unwise to rush to lock in solutions to problems that are not yet well understood in standards as rigorous as the Ada language. Moreover, real-time systems typically have inherent hardware-dependencies. Thus, there appears to be a need for more capability to economically adapt Ada language implementations to meet the special needs of real-time applications, based on existing Ada compilation technology and within the current Ada standard.